The present invention relates to an ignition method for an internal combustion engine, an injection being alternatively performed in at least one first operating mode or in a second operating mode, and the ignition coil being charged as a function of the current operating mode; and the present invention relates to a corresponding ignition device.
Although applicable to any fuels and engines of any vehicles, the present invention and the problem on which it is based are explained with reference to a direct gasoline-injection system of an engine of a passenger car.
FIG. 4 illustrates the dependence of torque M on engine speed N for different operating modes of an internal combustion engine.
During so-called homogeneous, normal operation H1 of the direct gasoline-injection system, the entire combustion chamber is homogeneously filled with a stoichiometric air-fuel mixture (lambda value xcex=1), which is ignited by the ignition sparks at the ignition firing point. In this case, there may be no ignition problems at all when the mixture has a high energy density.
However, homogeneous operation may also be realized in a lean manner and/or with exhaust-gas recirculation (EGR) as homogeneous operation H2. In this case, a high level of flow may be required in order to achieve sufficiently rapid burning in the case of low energy densities of the mixture in the combustion chamber. This may deflect the spark plasma, until it breaks away and reignition occurs.
In this manner, the spark energy during coil ignition may be distributed with typical spark durations of approximately 1 ms under these conditions, to numerous, subsequent sparks, which each reach new mixture regions.
But since the leanest operation or so-called high-EGR operation may only be attained when the entire energy of the ignition coil is introduced into a single flame core, all of the energy stored in the ignition coil may be required therefore to be supplied in such a short time that the spark still does not break away within this span of time (such as, for example, approximately 0.3-0.6 ms.).
This may yield a demand for as high an energy as possible and a very short spark duration (approximately 0.3-0.6 ms) for this H2 operation, which may result in a high, required initial current of 150-200 mA.
In order to make use of the fuel-consumption features with internal combustion engines having direct gasoline injection, so-called charge stratification may be implemented in the combustion chamber in certain operating ranges, which is referred to below as stratified-charge operation S.
During stratified-charge operation S, only a small, locally ignitable stoichiometric cloud is introduced into the combustion chamber, whereas the remaining contents of the combustion chamber may not be ignited. A feature of this stratified-charge operation S may include that the lean-combustion operation of the engine is extended, and fuel may therefore be saved in the end. Therefore, it may be desirable to configure the operating range of stratified-charge operation S to be as large as possible, and in particular, to therefore expand it to loads and engine speeds that are as high as possible.
During stratified-charge operation S, marked local and/or temporal lambda fluctuations may be present at the location of the ignition spark, when the average energy density in the mixture cloud is high. In order to achieve reliable ignition in this case, the spark should burn for a long time (such as, for example, approximately 5-10xc2x0 KW (KW=crank angle)), so that within this time, the formation of the flame core may be started when a flammable mixture region is seized by the spark plasma.
In this context, depending on the flow of the mixture at the spark plug, only a continuously decreasing portion of the electrical energy introduced from the ignition coil may be available for forming the flame core as the spark duration increases. Thus, the conventional proposal may generate a pulse train, i.e. to repeatedly charge and discharge the ignition coil, within the above-mentioned KW interval.
Therefore, an individual ignition spark that burns as along as possible with an initial current of, for example, approximately 50-80 mA and a secondary energy of, for example, approximately 80-100 mJ, or an adjustable-length pulse train with an initial current of, for example, approximately 100 mA from a coil having, for example, approximately 30 mJ of secondary energy, may be suitable for this stratified operating mode.
Since the demands for stratified S and homogeneous H1 and H2 operating ranges may therefore be markedly different, a conventional system configuration having individual sparks may create a conflict of aims, which may have previously only been approached as a compromise. An ignition coil may either be configured for a long spark duration (high secondary inductance, i.e. high number of secondary windings per unit length) with a moderate initial current, or for a short spark duration (low secondary inductance, i.e. low number of secondary windings per unit length). Therefore, a decision for a discrete configuration as a compromise may be essential.
In contrast to the conventional configuration approaches, an exemplary ignition method and/or exemplary ignition device of the present invention may provide that a functionality adapted to the problem of direct gasoline-injection engines may allow optimum ignition in stratified operation, as well as in homogeneous lean-combustion operation and/or with EGR, and in cold starting or other critical engine conditions.
The operating mode may be controlled as required. Only the amount of energy required for ignition may be introduced. This may prevent spark-plug wear.
A smaller space for the coil due to a smaller number of turns per unit length on the secondary side, or a larger iron cross section, may be provided in the same space. Therefore, a cost advantage may be attained by dispensing with the magnets for pre-magnetizing the iron circuit.
The type of ignition suitable for the specific operating mode may be provided by control-pulse coding. For example, a pulse-train ignition suitable for stratified operation may be combined with the option of loading the ignition coil with a markedly higher amount of energy during homogeneous operation by increasing the primary current, so that it still discharges as a single spark within the desired spark duration of approximately 0.3-0.6 ms.
According to a further exemplary refinement, the first operating mode may be a homogeneous, normal operation, which may be divided up into the submodes of stoichiometric normal operation and sub-stoichiometric normal operation, and the second operating mode may be an inhomogeneous stratified-charge operation.
According to a further exemplary refinement, the charging of the ignition coil during inhomogeneous, stratified-charge operation may be performed in the form of pulse-train ignition with a predetermined primary current, and the charging of the ignition coil during homogeneous operation may be performed in the form of a single-pulse ignition with an increase in the primary current.
According to a further exemplary refinement, the control-pulse curves characteristic of the current operating mode may have different pulse times and/or numbers of pulses. Thus, virtually all operating states may be coded, using a simple arrangement.
According to a further exemplary refinement, the iron circuit of the ignition coil may be controlled up to the start of saturation, in an operating mode that requires a high initial spark current. Thus, more energy may be stored and the rate of increase of the voltage may be increased because of the lower, secondary inductance at the beginning.